Welcome SC18 Attendees!
Nov. 9, 2018—We would like to welcome our fellow attendees of Supercomputing 2018 in Dallas, TX to our site. This page contains our presentation schedule as well as all our posters, presentation slides, relevant links, source code, and presenter bios. We will be in booth 3907 all week.
ACCRE research featured in PBS documentary “Neanderthal”; parts 1 and 2 online now
Oct. 11, 2018—Vanderbilt professor Tony Capra has been featured in “Neanderthal”, a two-part series on PBS that aired this month. His lab uses ACCRE extensively to explore the genetics of human evolution. Watch the first part of Neanderthal here (expires November 14). Watch the second part of Neanderthal here (expires November 21). Here is a clip from...
Problems with GPFS; logins and jobs may be affected
Nov. 7, 2018—Update, 3pm: /home is back online. Please check your jobs’ output very carefully as it is likely that many will need to be re-run, especially if they were performing I/O to or from /home. Jobs performing I/O to /scratch, /data, or /dors may have survived. Please open a helpdesk ticket with us if you have...
11/30/2018: Basics of Python Performance and Multithreading with Eric Appelt
Oct. 31, 2018—Our next Pizza and Programming Seminar will be held on Friday, Nov 30 beginning at 1:15pm in Stevenson Center 5326. Eric Appelt, Sr Sys Admin at ACCRE, will be leading an interactive seminar on the inner workings of Python. An abstract is included below. You can register for the seminar here: Make sure to register...
[Resolved] Cluster unresponsive or sluggish for some users
Oct. 18, 2018—Update, 2:30pm: All clear – if you notice anything usual, submit a helpdesk ticket as always. We’re receiving reports from users this morning about the cluster being unresponsive or sluggish. We are investigating the issue and will have an update soon. Thanks!
[Resolved] /scratch and /data partially offline
Oct. 11, 2018—Update, 10/16/2018: /scratch and /data are now 100% available again and we have successfully upgraded the cache in all of our GPFS storage appliances. Please open a helpdesk ticket with us if you have any questions or notice anything unusual. Update, 10/15/2018, 5pm: We are still working to get parts of /scratch and /data back...
[Resolved] Cluster inaccessible
Oct. 1, 2018—Update, 2pm: The cluster is accessible again. Thank you for bearing with us and we sorted through these issues. We were able to successfully upgrade the cache on three of our six GPFS storage appliances. The third one is where we encountered issues. A hardware problem prevented us from booting the device cleanly until an...
[Resolved] Network maintenance ongoing; some login issues and/or stale file handles possible
Sep. 27, 2018—Update, 10/1/2018: This has been resolved. We have been experiencing some network issues for the last few days that have impacted logins and caused occasional stale file handle errors. Later today, we will be performing some maintenance that we hope will resolve this issue or at least further isolate the root cause. This work (which...
[Resolved] Sporadic stale file handles & potential error messages when logging in to CentOS 7 environment
Sep. 18, 2018—Update 9/21/2018: It appears this issue has been resolved. Please open a Helpdesk ticket if you experience any problems. This morning we began sporadically seeing stale file handle messages in our CentOS 7 environment, which may have also impacted logins to login7.accre.vanderbilt.edu. We initially thought this due to a bad hard drive on one of...
CentOS 7 Upgrade: login.accre.vanderbilt.edu now points to CentOS 7, login6.accre.vanderbilt.edu available temporarily for GPU users
Aug. 13, 2018—Update, 9/28/2018: We have now pointed login.accre.vanderbilt.edu to the CentOS 7 environment. Update, 9/27/2018: Tomorrow afternoon we will be pointing login.accre.vanderbilt.edu at the CentOS 7 environment. Due to DNS caching this change may not take effect until Saturday or later on your local machine. After the switchover, if you are asked to verify that you trust...